Unlock 30% off on Manual Testing Annual Plans this Holiday Season.

Copied to Clipboard!

How to use c_test_1 method in Lemoncheesecake

Best Python code snippet using lemoncheesecake

NaiveBayesKde.py

Source:NaiveBayesKde.py

1import numpy as np2import matplotlib.pyplot as plt3from sklearn.neighbors import KernelDensity4from sklearn.model_selection import StratifiedKFold5class NaiveBayesKde:6    def __init__(self):7        self.best_bw = 08        self.best_valid_err = 19        self._trained = False10        self._errors = []11    def _kde_estimate(self, fit_x, train_x, valid_x, bw):12        kde = KernelDensity(bandwidth=bw).fit(fit_x)13        return kde.score_samples(train_x), kde.score_samples(valid_x)14    def _kde(self, fit_x, test_x, bw):15        kde = KernelDensity(bandwidth=bw).fit(fit_x)16        return kde.score_samples(test_x)17    def _kde_calc_fold(self, Xs, Ys, tr_ix, va_ix, bw):18        x_c0 = Xs[tr_ix[Ys[tr_ix] == 0]]19        x_c1 = Xs[tr_ix[Ys[tr_ix] == 1]]20        # probability of the classes21        c_train_0 = c_valid_0 = np.log(len(tr_ix[Ys[tr_ix] == 0]) / len(tr_ix))22        c_train_1 = c_valid_1 = np.log(len(tr_ix[Ys[tr_ix] == 1]) / len(tr_ix))23        for feat in range(0, Xs.shape[1]):24            # calculate logarithmic density for class 0, i.e log(P(x_i | c_0))25            t_dens, v_dens = self._kde_estimate(x_c0[:,[feat]], Xs[tr_ix][:,[feat]], Xs[va_ix][:,[feat]], bw)26            c_train_0 += t_dens27            c_valid_0 += v_dens28            # calculate logarithmic density for class 1, i.e log(P(x_i | c_1))29            t_dens, v_dens = self._kde_estimate(x_c1[:,[feat]], Xs[tr_ix][:,[feat]], Xs[va_ix][:,[feat]], bw)30            c_train_1 += t_dens31            c_valid_1 += v_dens32        # calculate predictions33        predict_train = np.argmax([c_train_0, c_train_1], axis=0)34        predict_valid = np.argmax([c_valid_0, c_valid_1], axis=0)35        # compare the predictions with our data sets36        cmp_train = Ys[tr_ix] == predict_train37        cmp_valid = Ys[va_ix] == predict_valid38        # obtain training error and validation error39        train_err = (len(tr_ix) - np.sum(cmp_train)) / len(tr_ix)40        valid_err = (len(va_ix) - np.sum(cmp_valid)) / len(va_ix) 41        return train_err, valid_err42    def optimize_bandwidth(self, X_r, Y_r, folds, start_bw, end_bw, step_bw):43        kfold = StratifiedKFold(n_splits=folds)44        best_bw = 045        best_valid_err = 146        for bw in np.arange(start_bw, end_bw + step_bw, step_bw):47            train_err = valid_err = 048            for tr_ix, va_ix in kfold.split(Y_r, Y_r):49                tr_err, va_err = self._kde_calc_fold(X_r, Y_r, tr_ix, va_ix, bw)50                train_err += tr_err / folds51                valid_err += va_err / folds52            self._errors.append([bw, train_err, valid_err])53            if (valid_err < best_valid_err):54                best_bw = bw55                best_valid_err = valid_err56        self.best_bw = best_bw57        self.best_valid_err = best_valid_err58        return self.best_bw, self.best_valid_err59    def fit(self, X_r, Y_r):60        if self.best_bw == 0:61            raise Exception('The bandwidth should be optimized first')62        self._p_c0 = np.log(np.sum(Y_r == 0) / len(Y_r))63        self._p_c1 = np.log(np.sum(Y_r == 1) / len(Y_r))64        self._x_c0 = X_r[Y_r == 0]65        self._x_c1 = X_r[Y_r == 1]66        self._trained = True67        68    def predict(self, X_t, bw):69        if self._trained == False:70            raise Exception('The classifier is not trained')71        c_test_0 = self._p_c072        c_test_1 = self._p_c173        for feat in range(0, X_t.shape[1]):74            test_dens = self._kde(self._x_c0[:,[feat]], X_t[:,[feat]], bw)75            c_test_0 += test_dens76            test_dens = self._kde(self._x_c1[:,[feat]], X_t[:,[feat]], bw)77            c_test_1 += test_dens78        predictions = np.argmax([c_test_0, c_test_1], axis=0)79        return predictions80    def plot_errors(self, save_fig=True, fig_name='NB.png'):81        errors = np.array(self._errors)82        plt.figure()83        plt.title('Training vs Cross Validation Errors')84        plt.plot(errors[:,0], errors[:,1], 'b', label='Training Error')85        plt.plot(errors[:,0], errors[:,2], 'r', label='Cross-Validation Error')86        plt.legend()87        88        if save_fig:89            plt.savefig(fig_name)90        plt.show()...

nankomah_Project2.py

Source:nankomah_Project2.py

1#!/usr/bin/env python32# -*- coding: utf-8 -*-3# import os4# os.chdir('./Project2/')5import numpy as np6from cvxopt import matrix7from cvxopt.solvers import qp8import matplotlib.pyplot as plt9def generate_vars(n=100, m = 30):10    """[returns B- nxn, A - mxn, c - mx1]11    Args:12        n (int, optional): [number of columns]. Defaults to 100.13        m (int, optional): [number of rows]. Defaults to 30.14    Returns:15        [np arrays]: [Generated Matrix and Vector]16    """    17    sigma = 1 / (4*n)18    I = np.eye(n)19    K = np.random.normal(scale=sigma, size=(n,n))20    21    A = np.random.uniform(size=(m,n))22    B = I + K + K.T23    c = np.random.uniform(size=(m))24    return A, B, c25def cvxpot_qp(A,c,B):26    """[return x_star]27    Args:28        A ([np.array]): [A - mxn]29        c ([np.array]): [c - mx1]30        B ([np.array]): [B - nxn]31    Returns:32        [type]: [x_star]33    """    34    m = A.shape[0]35    n = A.shape[1]36    q = matrix(np.zeros((n,1)))37    P = matrix(B)38    A = matrix(A)39    b = matrix(c)40    cvx_results= qp(P = P, q = q, A=A, b=b)41    return np.asarray(cvx_results['x']).flatten()42def lagrangian_x(A,c,B,c_k,lam):43    """[returns x_k based on d_x alm solve for x   ]44    Args:45        A ([np.array]): [A - mxn]46        c ([np.array]): [c - mx1]47        B ([np.array]): [B - nxn]48        c_k ([float]): [variable will the following condition49         must be > 0, -> infty, c_{i} <= c_{i+1} ]50        lam ([np.array]): [lam - mx1]51    Returns:52        [x]: [mx1 np.array]53    """    54    leftside = 2*B+(c_k*A.T@A)55    rightside = (c_k*A.T@c) - (A.T@lam)56    x = np.linalg.pinv(leftside)@rightside57    return x.flatten()58def constraint(A,x,c):59    """[returns output of constraint]60    Args:61        A ([np.array]): [A - mxn]62        x ([np.array]): [x - mx1]63        c ([np.array]): [c - mx1]64    Returns:65        [type]: [mx1]66    """    67    return (A@x) - c 68def alm(B,A,c, epsilon=10**(-10)):69    """[uses Augmented Lagrangian method to find an x_k simlar to x_star]70    Args:71        B ([np.array]): [B - nxn]72        A ([np.array]): [A - mxn]73        c ([np.array]): [c - mx1]74        epsilon ([float], optional): [threshold value]. Defaults to 10**(-13).75    Returns:76        [list]: [list of relative error ]77    """    78    x_star = np.asarray(cvxpot_qp(A,c,B*2)).flatten()79    error_i = [] 80    lambda_k = np.zeros(c.shape)81    #c_k must be > 0, -> infinity, c_k_{i} <= c_k_{i+1} 82    c_k =.183    i = 0 84    while True:85        x_k = lagrangian_x(A,c,B,c_k,lambda_k)86        error = np.linalg.norm(x_k - x_star) /  np.linalg.norm(x_star)87        error_i.append(error)88        stopping = np.linalg.norm(A@x_k - c)89        if stopping <= epsilon:90            return x_k, i, error_i91        92        lambda_k += c_k*constraint(A,x_k,c)93        c_k = 2.0**i94        i += 195def main():96    """[main function]97    """    98    #a)99    A = np.loadtxt('A_test_1.txt')100    c = np.loadtxt('c_test_1.txt')101    B = np.loadtxt('B_test_1.txt')102    # Q, A, b = generate_vars()103    #b)104    x_star = cvxpot_qp(A,c,B*2)105    print('cvx output')106    print(x_star)107    #c) 108    x_k, i, error_i = alm(B,A,c)109    print()110    print('Augmented Lagrangian method output')111    print(x_k)112    print()113    print('Final Error')114    print(error_i[-1])115    print()116    print('Final Error is below 10^(-6)')117    print(error_i[-1] <= 10**(-6))118    print()119    120    fig = plt.figure()121    plt.title('Relative Error')122    plt.plot(np.arange(len(error_i)), error_i, label = 'Relative Error vs k')123    plt.legend()124    plt.grid(True)125    fig.tight_layout()126    # plt.savefig('pix/Given_sample.png', dpi=fig.dpi)127    plt.show()128if __name__ == "__main__":...

B.py

Source:B.py

...5}6@lcc.suite("BB1")7class BB1:8    @lcc.test("Test of BB1")9    def c_test_1(self):10        import time11        time.sleep(0.21)12        require_that("value", 1, is_integer(2))13@lcc.suite("BB2")14class BB2:15    @lcc.test("Test of BB2")16    def d_test_1(self):...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.